Construction of a cDNA library encoding pea seed proteins

Abstract

A cDNA library was constructed using mRNA isolated from the cotyledons of developing seeds of pea (Pisum sativum L.) 13-17 days after flowering, the mid-stage of development. The library was systematically screened for clones encoding most of the more abundant seed proteins. Representative clones of each class were isolated and characterised by restriction mapping, hybridisation experiments and in some cases DNA sequencing. The abundances of the various classes of clones were correlated with the known abundance classes of mRNA sequences at this stage of development and shown to fit well. Two types of cDNA clone corresponding to legumin were found. One type was found to have similar restriction maps to legumin genes A, B and C which encode "major" legumin polypeptides. A representative of the other type of legumin clone was shown to correspond to legumin gene J which encodes a "minor" legumin polypeptide. Restriction mapping was used to show that genes in this subfamily are more diverse than in the Leg A,B,C subfamily. Clones for the three main vicilin types were found. From their abundances and cross hybridisations, it was shown that the three vicilins are roughly equally homologous to each other and that their mRNAs are roughly equally abundant at this stage of development. A clone was found which encodes one of the major albumins and was completely DNA sequenced. The predicted protein sequence thus obtained was shown to match well with the partial protein sequences previously obtained from total major albumins. Clones for the soluble seed lectin were found and the DNA sequence of one was shown to be identical with the published sequence of pea lectin mRNA. However its poly A tail started 24 bases further downstream, after a series of overlapping sequences closely homologous to the consensus polyadenylation signal. The size of the lectin mRNA was found to be ~1000 bases, confirming the near completeness of the published sequence.